Article Text

other Versions


Critically ill children did not benefit from tight glycaemic control
  1. Angela Katrina Lucas-Herald1,
  2. Kenneth J Robertson2
  1. 1Developmental Endocrinology Research Group, University of Glasgow, Glasgow, UK
  2. 2Royal Hospital for Children, Glasgow, UK
  1. Correspondence to Dr Angela Katrina Lucas-Herald, Developmental Endocrinology Research Group, University of Glasgow, 1345 Govan Road, Glasgow G51 4TF, UK; angela.lucas-herald{at}

Statistics from

Study design

Design: Randomised controlled trial.

Allocation: Computer generated block design.

Study question

Setting: Thirty-five paediatric intensive care unit (ICU) centres.

Patients: Three hundred and sixty patients (median age 5.5 years, IQR 1.4–12.5 years) in the lower target group; 353 patients (median age 6.7 years, IQR 1.7–12.8 years) in the higher target group.

Intervention: Randomisation to low target glycaemic control group (80–110 mg/dL (4.4–6.1 mmol/L)) or higher target glycaemic control group (150–180 mg/dL (8.3–10 mmol/L).

Outcomes: Primary outcome: number of ICU-free days to day 28. Prespecified secondary outcomes: 90-day mortality, severity of organ dysfunction, the number of ventilator-free days to day 28, incidence of healthcare-associated infection and incidence of hypoglycaemia.

Patient follow-up: Patients were followed up for 90 days. Recruitment stopped at 50% due to low likelihood of benefit and potential of harm.

Main results

The key results from this study are summarised in Table 1. There was no change in 90 day hospital mortality or median number of ICU-free days in each group. There was however an increased incidence of healthcare associated infection and hypoglycaemia in the low target group.

VariableLow target
Higher target (n=349)p Value
Median (IQR), no of ICU-free days through day 2820 
90-day hospital mortality (%)52 (14.9)40 (11.5)0.22
Incidence of healthcare-associated infection12 (3.4)4 (1.1)0.04
Catheter-associated bloodstream infection
(Total no/no of days with CVC)
Catheter-associated UTI
(Total no/no of days with UTI)
Ventilator-associated pneumonia
(Total no/no of days with ventilator)
Hypoglycaemia (%)79 (22.6)33 (9.5)<0.001
  • CVC, central venous catheter; UTI, urinary tract infection.

  • Table 1. Summary of main results from the study.


Maintaining tight glycaemic control (80–110 mg/dL, 4.4–6.1 mmol/L) for critically ill children with hyperglycaemia in paediatric ICU did not infer any benefit in terms of number of ICU-free days or mortality; severity of organ dysfunction or number of ventilator-free days. Infection rates were increased.

Abstracted from

Agus MSD, Wypji D, Hirshberg EL, et al. Tight glycaemic control in critically ill children. New Engl J Med 2017 [Epub ahead of print, 24 Jan  2017].


The need to maintain tight glycaemic control in critically unwell patients has been a hot topic in intensive care medicine in recent years. One single-centre randomised controlled trial in adult surgical patients demonstrated a reduction in morbidity and mortality (van den Berghe et al, 2001), but this result was not shown consistently in other trials. It is known that there is a wide variation in practice in paediatric intensive care units; with the advent of continuous glucose monitoring it is now easier than ever before to track blood sugar readings and adjust insulin or fluid therapy as required. This study was therefore important and necessary in the effort to improve care and outcomes for critically ill children.

The study demonstrates that there was no benefit in the tight glycaemic control of critically ill children and that morbidity and 90-day mortality were increased in the low-target group. A rise in blood glucose is a normal physiological response to critical illness and a forced reduction in this may therefore do more harm than good. However, the increased risk secondary to the maintenance of tight glycaemic control may represent the administration of insulin therapy with its other metabolic effects as opposed to the absolute blood glucose levels.

The finding that there was an increase in healthcare-associated infections should be noted. Of course, this study was not powered to detect this effect and may reflect the sample size rather than a true finding.

In conclusion, this study shows that there is no clinical benefit for maintaining blood glucose control between 80 and 110 mg/dL (4.4–6.1 mmol/L) in critically unwell children. This is of relevance to any paediatricians caring for such children. Guideline development to standardise management for affected children would be useful.

View Abstract


  • Contributors AKL-H and KJR contributed equally to the preparation of the submitted manuscript.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.